Abstract. The European Centre for Medium-Range Weather Forecasts (ECMWF) recently
released the first 7-year segment of its latest atmospheric reanalysis: ERA-5
over the period 2010–2016. ERA-5 has important changes relative to the former ERA-Interim
atmospheric reanalysis including higher spatial and temporal resolutions as
well as a more recent model and data assimilation system. ERA-5 is foreseen
to replace ERA-Interim reanalysis and one of the main goals of this study is
to assess whether ERA-5 can enhance the simulation performances with respect
to ERA-Interim when it is used to force a land surface model (LSM). To that
end, both ERA-5 and ERA-Interim are used to force the ISBA (Interactions
between Soil, Biosphere, and Atmosphere) LSM fully coupled with the Total
Runoff Integrating Pathways (TRIP) scheme adapted for the CNRM (Centre
National de Recherches Météorologiques) continental hydrological
system within the SURFEX (SURFace Externalisée) modelling platform of
Météo-France. Simulations cover the 2010–2016 period at half a
degree spatial resolution. The ERA-5 impact on ISBA LSM relative to ERA-Interim is evaluated using
remote sensing and in situ observations covering a substantial part of the
land surface storage and fluxes over the continental US domain.
The remote sensing observations include (i) satellite-driven model
estimates of land evapotranspiration, (ii) upscaled ground-based observations
of gross primary production, (iii) satellite-derived estimates of surface
soil moisture and (iv) satellite-derived estimates of leaf area index (LAI).
The in situ observations cover (i) soil moisture, (ii) turbulent heat fluxes,
(iii) river discharges and (iv) snow depth. ERA-5 leads to a consistent
improvement over ERA-Interim as verified by the use of these eight
independent observations of different land status and of the model
simulations forced by ERA-5 when compared with ERA-Interim. This is
particularly evident for the land surface variables linked to the terrestrial
hydrological cycle, while variables linked to vegetation are less impacted.
Results also indicate that while precipitation provides, to a large extent,
improvements in surface fields (e.g. large improvement in the representation
of river discharge and snow depth), the other atmospheric variables play an
important role, contributing to the overall improvements. These results
highlight the importance of enhanced meteorological forcing quality provided
by the new ERA-5 reanalysis, which will pave the way for a new generation of
land-surface developments and applications.
Optimal averaging of soil moisture predictions from ensemble land surface model simulations
